Model searching

Performs Wald or score tests

modelsearch(x, k = 1, dir = "forward", type = "all", ...)

Arguments

x

lvmfit-object

k

Number of parameters to test simultaneously. For equivalence the number of additional associations to be added instead of rel.

dir

Direction to do model search. "forward" := add associations/arrows to model/graph (score tests), "backward" := remove associations/arrows from model/graph (wald test)

type

If equal to 'correlation' only consider score tests for covariance parameters. If equal to 'regression' go through direct effects only (default 'all' is to do both)

...

Additional arguments to be passed to the low level functions

Value

Matrix of test-statistics and p-values

See also

compare, equivalence

Author

Klaus K. Holst

Examples

m <- lvm();
regression(m) <- c(y1,y2,y3) ~ eta; latent(m) <- ~eta
regression(m) <- eta ~ x
m0 <- m; regression(m0) <- y2 ~ x
dd <- sim(m0,100)[,manifest(m0)]
e <- estimate(m,dd);
modelsearch(e,messages=0)
#>  Score: S P(S>s) Index  holm BH    
#>  0.0734   0.7864 y3~~x  1    0.7864
#>  0.0734   0.7864 y3~x   1    0.7864
#>  0.0734   0.7864 x~y3   1    0.7864
#>  0.0734   0.7864 y1~~y2 1    0.7864
#>  0.0734   0.7864 y1~y2  1    0.7864
#>  0.0734   0.7864 y2~y1  1    0.7864
#>  0.1991   0.6554 y2~~y3 1    0.7864
#>  0.1991   0.6554 y2~y3  1    0.7864
#>  0.1991   0.6554 y3~y2  1    0.7864
#>  0.1991   0.6554 y1~~x  1    0.7864
#>  0.1991   0.6554 y1~x   1    0.7864
#>  0.1991   0.6554 x~y1   1    0.7864
#>  0.675    0.4113 y1~~y3 1    0.7864
#>  0.675    0.4113 y1~y3  1    0.7864
#>  0.675    0.4113 y3~y1  1    0.7864
#>  0.675    0.4113 y2~~x  1    0.7864
#>  0.675    0.4113 y2~x   1    0.7864
#>  0.675    0.4113 x~y2   1    0.7864
modelsearch(e,messages=0,type="cor")
#>  Score: S P(S>s) Index  holm BH    
#>  0.0734   0.7864 y3~~x  1    0.7864
#>  0.0734   0.7864 y1~~y2 1    0.7864
#>  0.1991   0.6554 y2~~y3 1    0.7864
#>  0.1991   0.6554 y1~~x  1    0.7864
#>  0.675    0.4113 y1~~y3 1    0.7864
#>  0.675    0.4113 y2~~x  1    0.7864